The linear no-threshold (LNT) model has been widely used to establish international rules and standards in radiation protection. It is based on the notion that the physical energy deposition of ionizing radiation (IR) increases carcinogenic risk linearly with increasing dose (i.e., the carcinogenic effectiveness remains constant irrespective of dose) and, within a factor of two, also with dose-rate. However, recent findings have strongly put into question the LNT concept and its scientific validity, especially for very low doses and dose-rates. Low-dose effects are more difficult to ascertain than high-dose effects. Epidemiological studies usually lack sufficient statistical power to determine health risks from very low-dose exposures. In this situation, studies of the fundamental mechanisms involved help to understand and assess short- and long-term effects of low-dose IR and to evaluate low-dose radiation risks. Several lines of evidence demonstrate that low-dose and low dose-rate effects are generally lower than expected from high-dose exposures. DNA damage signaling, cell cycle checkpoint activation, DNA repair, gene and protein expression, apoptosis, and cell transformation differ qualitatively and quantitatively at high- and low-dose IR exposures, and most animal and epidemiological data support this conclusion. Thus, LNT appears to be scientifically invalid in the low-dose range.